Cancer Browser - Selection

Cancer Browser Tutorial

The Cancer Browser provides an easy method to explore the mutation spectrum of a cancer phenotype (eg Small Cell Lung Carcinoma). The initial page display loads with 4 panels, showing an alphabetically ordered list of primary tissues in the first panel and three empty panels for Sub Tissues, Histology and Sub Histology. Selecting a primary tissue will offer further options for selecting a sub tissue and its phenotype morphology (histology and sub histology), for example: Lung » 'Include All' » Carcinoma » Small cell carcinoma.

Each tissue, sub tissue, histology and sub histology is listed with counts of mutated and analyzed samples adjacent to it in the brackets.

Click "Go" at any stage of the selection process to view a chart of the top mutated genes listed in order of mutation frequency, together with a range of tabs presenting other views on the data selected.

For more details on counts please follow this link.

Note: After making any changes in the selection process , press "Go" again to reload the data.

Cancer Browser - Top 20 Genes

This section of the Cancer Browser page presents a list of genes with statistics to show how many tumours have been examined in each gene and the tumour type's mutation frequency.

This chart shows the top 20 mutated genes for the selected tissue/disease. The mutation frequencies are displayed next to the gene name. The chart is ordered by the number of mutated samples examined. Where two or more genes have the same number of mutated samples, we then order by the total number of samples.

This ordering system is used in preference to ordering by frequency because the latter method artificially inflates the significance of genes with low sample counts. For example, a gene with 10 samples 9 of which are mutated would be ranked above a gene with 10000 samples of which 1000 are mutated.

By default this graph is drawn using known cancer genes listed in the Cancer Gene Census. The graph can be toggled to show all genes in COSMIC or only known cancer genes (census) genes by clicking the link under the graph. The graph can also be redrawn to display only samples with mutations or only analysed samples by clicking the red or blue key boxes at the bottom of graph.

The interactive bar chart displays Gene Names with mutation frequencies (total mutated samples / total samples analysed). The blue bar shows the mutated samples and the red bar shows the total number of samples analysed; mouseover the bars for sample counts. Note that the samples analysed also includes samples tested for fusions although samples with fusions can only be viewed in the Variants data /Fusions tab further down the page. Click on any of the bars or the gene name to explore the mutation information in more detail.

Cancer Browser - Mutation Matrix

This tab shows a "Mutation Matrix" plot between genes and samples for the selected tissue/phenotype. The chart contains a maximum of 20 genes (rows) and 175 samples (columns) with each box representing a Gene-Sample combination.

Each box is colour coded with purple representing Point mutations, orange representing CNV gain and blue representing CNV Loss. For aberrant gene expression a red border indicates over expression or a green border under expression. When samples are positive for multiple categories colour codes will be combined eg for CNV gain and Over expression the box will be orange with a red border.

Gene ranking:

Genes are ranked based on the total count of samples having an aberration from the chosen category (highest at the top). Each category is explained below.

All - All annotations ( Point mutations/ CNVs/ Gene expression )
Point - Point mutations
Gain - Copy number gain
Loss - Copy number loss
Over expression - Over expressed gene
Under expression - Under expressed gene.

Sample ranking:

Samples are ranked based on the count of aberrations in the top 20 ranked genes. The more genes with aberrations the higher the rank (highest on the left). Multiple annotations in the same category are equally weighted e.g a gene having 10 point mutations, 1 CNV gain, and 1 Over expression will be counted as 1 in categories All, Point, CNV Gain and Over expression, but counted as 0 in categories CNV Loss and Under expression.

Cancer Browser - Mutant Samples

This table shows the sample name and mutation count for the tissue/disease selected.

Cancer Browser - Non-Mutant Samples

This table shows the sample name and ID where no mutations were detected for the tissue/disease selected.

Cancer Browser - Mutations

This tab shows a table of mutations found for the selected disease. This table also provides the ability to search, sort and export the table in csv or tsv format.

Gene Name - Gene name in which the mutation has been found. It links to the gene analysis page for more details on that gene.
Samples - Sample name in which the mutation has been found. It links to the sample overview page for more details on that sample.
CDS mutation - Details of the change that has occurred in the nucleotide sequence. Formatting is based on the recommendations made by the Human Genome Variation Society. The description of each type can be found by following the link to the Mutation Overview page.
AA Mutation - Details of the change that has occurred in the peptide sequence. Formatting is identical to the method used for the nucleotide sequence.

Cancer Browser - Fusion

This tab shows a table of fusions found for the selected disease.

Gene Pair - These are the genes involved in fusions and it links to the fusion overview page, for more details on fusions please click here.
Fusion - Fusion mutations identified for the specified gene pair in this disease, linking to the fusion summary page. For more details on fusions please click here.
Positive Samples - The number of unique samples with mutations for the specified fusion gene pair.

Cancer Browser - Genes With Mutations

This tab contains a list of all the genes with mutated samples for the selected disease. Each gene name is linked to the gene anlaysis page showing the mutation distribution histogram, maintaining the filter for the selected disease. This table also provides the ability to search, sort and export the data in csv and tsv format.

Gene - Lists all the genes screened with mutations, it links to the gene analysis page with the disease selection already provided in the above panel.
Mutated Samples - Number of unique mutated samples found in the this gene for the selected disease.
Samples Tested - Total number of unique samples tested for the specified gene in the selected disease.

For more details on counts please follow this link .

Cancer Browser - Genes Without Mutations

This tab contains a list of all the genes with no mutations for the selected disease. This table also provides the ability to search, sort and export the data in csv and tsv format.

Gene - Lists all the genes screened with no mutations, it links to the gene analysis page with the disease selection already provided in the above panel.
Mutated Samples - Number of unique mutated samples found in the this gene for the selected disease.
Samples Tested - Total number of unique samples tested for the specified gene in selected disease.

For more details on counts please follow this link .

Cancer Browser - Methylation

This tab contains a list of all the genes/probes where there is differential methylation for the selected disease. The 'Low', High' and 'Tested' columns show the number of samples.This table also provides the ability to search, sort and export the data in csv and tsv format.

Gene - Lists the gene associated with the probe/target CpG or '-' if none (eg the probe maps to an intron, UTR or promoter).
Probe - the unique identier of the probe.
Position - the genomic location of the probe (with links to the COSMIC Cancer Genome Browser and Ensembl).
Methylation Low - The number of samples where the beta-value is <0.2 and the methylation state is altered relative to the normal average.
Methylation High - The number of samples where the beta-value is >0.8 and the methylation state is altered relative to the normal average.
Methylation Tested - The total number of tumour samples tested.

For more details on counts please follow this link.

For more details on methylation data analysis and definitions used, please follow this link.

Cancer Browser - CNV & Expression

This tab shows a table of all the genes present in COSMIC and the number of samples matching the tissue/histology selections with copy number variants (CNV gain/loss), and gene over/under expression.

Only genes which have CNV and/or aberrant gene expression are shown.

The table contains the following columns -

Gene - Cosmic gene name ( All genes are shown )
Copy Number
- Gain - Total number of samples which have gain
- Loss - Total number of samples which have loss
- Tested - Total number of samples analysed for this data type
Expression
- Over - Total number of samples where the gene is over expressed
- Under - Total number of samples where the gene is under expressed
- Tested - Total number of samples analysed for this data type

Cancer Browser - Distribution

This tab shows mutation distribution pie charts and histograms for the selected disease.

Overview Chart

This interactive pie chart displays an overview of the mutation spectrum for the selected disease, mouseover the pie chart or table columns to highlight regions of interest. Mutation data can be exported from this table by clicking the 'Mutation Type' links.

Colour - The colours in the pie chart are matched to the colours in the table.
Mutation Type - All the mutations are classified as either substitution, deletion, insertion or complex mutations, the 'other' category contains all those mutations that fall outside the defined categories or if there is no information for nucleotide changes. This links to a page detailing much additional information on the relevant samples and mutations.
Mutant Samples - Number of unique mutated samples for the selected disease. Each mutation is counted, each sample is only counted once in the Total. Therefore if a sample has for instance, both missense and nonsense mutations, it will contribute to both of these counts, but only once to the Total count; as a result of this, the total count will often be a slightly lower number than the sum of the individual mutation counts.
Percentage - Observed frequency of this mutation type.

Note that samples may have multiple mutations, which can fall into different categories. For example, one sample may have a nonsense mutation and a synonymous mutation, and when thinking about the percentages in the table, that sample is "double counted" in a sense.

For each type of mutation, we calculate the percentage as num_samples / total samples * 100. For example, for nonsense mutations in lung cancer samples, we get 1007 / 34426 * 100 = 2.93%. For synonymous mutations, we get 816 / 34426 * 100 = 2.37%. However, those two values don't take into account the possibility of the same sample appearing in both calculations, and if we simply add them together we'll get an over-inflated percentage score — the total will be more than 100%.

Substitutions (coding strand)

This pie chart gives a detailed overview of substitution mutations for the base pair changes on the coding strand. It captures all the changes from A,C,G,T to A,C,G,T. If there is no nucleotide information available then it will be reported separately at the bottom of the graph.

Substitutions (both strands)

This subsitution mutations pie chart provides the genomic overview irrespective of the coding strand. This pie chart is based on nucleotide base pair changes C:G>A:T, C:G>G:C , C:G>T:A , T:A>A:T, T:A>C:G, T:A>G:C. If there is no nucleotide information available then it will be reported separately at the bottom of the graph.

Deletions

This bar graph shows the number of deletions in the selected disease on the X axis with the nucleotide length of deletions on the Y axis. Click on the blue bar to see additional details, if there is no nucleotide information available then it will be reported separately at the bottom of the graph.

Insertions

This bar graph shows the number of insertions in the selected disease on the X axis with the nucleotide length of insertions on the Y axis. Click on the blue bar to see additional details, if there is no nucleotide information available then it will be reported separately at the bottom of the graph.